Abstract
At a glance, the relationship between my technical work and my STS research may seem non-existent. This is largely due to the fact that capstone projects are assigned based on preferences for systems engineering majors. Although healthcare is a captivating area of study, I knew I did not want to spend a year working on two projects within the same industry. Consequently, I choose to focus my STS research on an industry I interact with almost daily: online thrifting. Despite the difference in work, both projects are connected through a shared emphasis on optimization and efficiency. My capstone project is centered around improving hospital scheduling efficiency, while my STS research explores how optimization impacts outcomes across the socio-technical system that is online thrifting platforms. The connection between the projects deepens in how they both treat optimization not just as a technical objective, but as a value built into a system design. While my capstone demonstrates the benefits of efficiency in healthcare, my STS research shows its consequences in consumer platforms with conflicting priorities. Taken together, these projects pushed me to see optimization not as inherently beneficial, but instead as a value-laden choice whose impact depends on the goals and interests driving it.
Surgery is a cornerstone of modern healthcare, with the operating room (OR) being one of the most resource intensive units within a hospital. Since surgeries account for a substantial portion of hospital revenue, accurate prediction of surgery durations and efficient OR scheduling are critical to minimize delays for patients, as well as to reduce staff overtime and OR idle time. However, OR scheduling is complex due to variability including patient history, unforeseen complications, and emergency cases that can disrupt planned schedules. This complexity is compounded by the current OR scheduling approach, which remains a highly manual process driven largely by human decision-making and simple statistical averages. The goal of my team’s capstone project was to develop a model that improves elective surgery duration predictions by identifying patterns from historical data provided by the UVA Department of Surgery. Given the large number of unique surgeries, data was clustered to gain greater insights into underlying patterns. Using this processed data, we developed XGBoost and Random Forest models that identify the most influential variables affecting surgery duration and improve time estimates for future procedures. Preliminary models showed a nearly 40-minute improvement in Root Mean Squared Error (RMSE) for time prediction, a 59% decrease from the current prediction method.
My STS research investigates an application of optimization in a different corner of modern technology. In recent years, thrifting platforms like Depop have emerged as major players in the online fashion industry by positioning secondhand shopping as a more sustainable and ethical alternative to fast fashion. However, through a comparative platform analysis of Depop and Shein, my analysis reveals that Depop's platform design shares substantial similarities with the fast fashion platforms Depop opposes. Both platforms use personalized feeds, infinite scroll, trend curation, and urgency mechanics in hopes of optimizing the browsing into purchasing rates. After observing these similarities alongside respective brand messaging, it is clear the primary difference lies not in structure, but in the framing. Shein emphasizes affordability and accessibility, while Depop emphasizes its sustainability and environmental impact. Applying the Social Construction of Technology (SCOT) framework, my research also argues that this convergence reflects design processes shaped by competing interests, which ultimately settle around a singular goal that accelerates consumption. My STS research shows how optimization logic, when driven by profit rather than stated mission, can quietly undermine a platform's core values and the quality of its sustainability claims.
Working on both these projects simultaneously was valuable because it led me to view optimization from both a technical and a big picture perspective at the same time. If I had completed each project in isolation, I would have been more likely to treat optimization in my capstone as strictly a performance goal (accuracy and efficiency), without questioning the broader implications of it. Conversely, my STS research may have remained more theoretical, critiquing platform design without truly understanding why optimization was implemented and in real systems. Exploring optimization across two unrelated fields also helped me reinforce the idea that the outcomes of efficiency-driven systems depend heavily on the values, interests, and goals of the people behind them. All in all, the two projects tell a more complete story about systems thinking, showing optimization's potential with its shortcomings. As a systems engineer, this is a distinction I hope to carry forward. Developing technology well means not only optimizing for performance but being intentional about the people and purposes that the design is meant to serve.
